A landmark regulatory approval has vaulted the cell and gene therapy space to the front of the drug industry's mind. But speedy progress has also brought forward significant challenges.
Approved for infants with spinal muscular atrophy, Zolgensma is a stunning advancement for a disease that in its most severe form is fatal before age two. In clinical tests, almost all babies given the therapy remained alive and off permanent breathing support.
"The issue is really how do you cover [Zolgensma], not how you say no to it," said Michael Sherman, the chief medical officer for Harvard Pilgrim Health Care, in an interview with BioPharma Dive on the sidelines of the 2019 BIO conference in Philadelphia. A similar logic applies to Spark Therapeutics' Luxturna, a gene therapy approved in the U.S. in December 2017 to treat a rare form of blindness.
Novartis' Zolgensma and Spark's Luxturna are just the first in what's expected to be a wave of experimental gene therapies to make it to market. The Food and Drug Administration foresees approving 10 to 20 cell and gene therapy products each year by 2025.
Not all of those products, though, are guaranteed to have the compelling efficacy that make Luxturna and Zolgensma relatively easy choices for insurers to cover.
The next critical test of the gene therapy era will likely come in hemophilia. Multiple companies are racing to market with candidates for hemophilia A, with BioMarin Pharmaceutical the furthest along.
"Hemophilia A is going to force the problem [with payers] to be solved," said Sandy Macrae, CEO of Sangamo Therapeutics, which is developing a rival gene therapy to BioMarin's.
"If it's SMA, it's a few dreadfully impacted children," he said. "No one's going to stand in the way of that, and they're just going to make it available. If it's hemophilia A, there are alternatives."
Payer pains in adapting to the new era
Novartis priced Zolgensma at $2.1 million, making it the most expensive drug ever brought to market.
The Swiss pharma intends for that cost to be spread over multi-year payment plans, the result of lengthy discussions with payers.
New England-based Harvard Pilgrim Health Care, for instance, began talks on Zolgensma more than a year ago, before Novartis bought the therapy's original developer, AveXis.
Still, the payment plan is less substantive than both sides had hoped for, Sherman said, reflecting limits within the U.S. healthcare system in paying for these one-time treatments.
Medicaid "best price" regulations enacted over 25 years ago require that state Medicaid agencies receive statutory discounts, either at a fixed rate or, if greater, matching what drugmakers provide elsewhere. That's proved a roadblock in constructing outcomes-based or annuity-based agreements with payers.
"I want to stress that the issue wasn't Novartis' lack of willingness to participate," Sherman said. "It was the best pricing regulations that have been the largest challenge."
"As someone who is trying to operationalize this on the front lines," he added, "the Medicaid best price restriction is the largest barrier to some of the innovative kind of agreements that both the pharma companies and myself would like to engage in."
Sherman said he's had similar talks with other gene therapy developers, including BioMarin, Spark, Sarepta Therapeutics and Bluebird bio that have all been predicated on that barrier being removed.
In hemophilia, Sherman noted that having existing treatments gives insurers the ability to say no, and uncertainty over the durability of gene therapy's benefits makes outcomes-based agreements even more critical.
Recent three-year data from BioMarin, for example, appeared to show a continued drop in clotting factor expression, albeit to levels considered "mild" hemophilia.
Gene therapy's rapid emergence may bump up against other limits, too.
Most treatments now being developed use inactivated viral vectors, such as adeno-associated viruses or lentiviruses, to deliver corrected genes. Those vectors can concentrate in the liver, potentially narrowing the universe of readily targetable diseases that also carry commercial appeal, Sangamo's CEO predicted.
"I think in the next three to five years, gene therapy will have mined all of the liver diseases that are liver-centric and really saturate that space," Macrae said.
Peter Marks, the director of the FDA's Center for Biologics Evaluation and Research, noted gene therapy will face additional challenges in addressing diseases spurred by multiple genes rather than the single gene defects that treatments like Luxturna and Zolgensma target.
"The challenge for gene therapies in areas where we don't understand the pathophysiology is that becomes a much larger development program, much more expensive," Marks said in an interview at BIO. "You lose the benefit you normally get with gene therapy, which is that you have a high probability of success if you're dealing with a single-gene disorder."
Regulatory review for the cell and gene therapies brought forward have so far heavily focused on what's known as chemistry, manufacturing, and controls — the details of how a therapy is made. Less focus has been on the clinical questions of how the treatment works, since single gene defects cause clearly identifiable deficiencies.
That's made interpreting clinical results from these first gene therapies like Zolgensma pretty simple.
"How many kids do you need to see that would be floppy babies on ventilators that instead are walking around at age three?" Marks said. "You don't need tons of those, and you don't need sophisticated statistics, patient-focused outcomes, because they're normal. You go from very abnormal to normal."
More uncertain mechanisms of action would add back in the "biological" risk of many current experimental drugs, which are often supported by hazy notions of how they work.
"The worry is that you're going to devolve back to conventional drug development, where clinical data will actually be important, because you'll have to figure out 'Is that actually really working' and 'Is it working over the long term?'" Marks said.
Some biotechs are already going after those tougher, multigenic targets. Marks highlighted Parkinson's disease as one such example. Axovant and Voyager Therapeutics are both developing gene therapies for that neurodegenerative disorder.
In 10 or 15 years, the CBER head suspects the gene therapy space may end up looking quite like the small molecule industry of today.
FDA keeping pace, with caution
Even as gene therapy companies bolt ahead, the head of CBER remains cautious of leading the agency too far forward in "leaps of faith with using preclinical data."
While recent years have been marked with success, gene therapy's history has its share of setbacks.
"We don't want to have another Jesse Gelsinger," Marks said, referring to the U.S. teenager who died in 1999 after being treated with an experimental gene therapy. "We don't want to go backwards."
"The good news is people have more understanding now for this, but if you look at what happened with some of the Phase 1 trials that went bad in Europe, it doesn't take much to get people to really pull back," he added.
Caution, though, is balanced against the FDA's interest in promoting gene therapy development, even when it's for just a few patients.
The regulator is considering new ideas for regulatory review of so-called bespoke, or extremely personalized therapies, Marks said. If there's a level of similarity, the FDA is looking for ways to streamline oversight so each tiny tweak won't require a different drug application.
"If you treated those each as individual drugs, people are never going to get the therapies because you can't expect anyone to put in the effort," he said. "And so you'll never have any business venture going after making these."
"We want to be forward-leaning," Marks added, but "we don't want to be cavalier and create problems."